SrTiO3 (111) Interface and Its Effect on

Feb 13, 2017 - High-temperature annealing in air is used to produce SrTiO3 (111) surfaces with two types of atomically flat terraces: those that promo...
0 downloads 0 Views 2MB Size
Subscriber access provided by University of Newcastle, Australia

Article 2

3

Buried charge at the TiO/SrTiO (111) interface and its effect on photochemical reactivity Yisi Zhu, Paul A. Salvador, and Gregory S. Rohrer ACS Appl. Mater. Interfaces, Just Accepted Manuscript • DOI: 10.1021/acsami.6b16443 • Publication Date (Web): 13 Feb 2017 Downloaded from http://pubs.acs.org on February 19, 2017

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

ACS Applied Materials & Interfaces is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 27

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

Buried charge at the TiO2/SrTiO3 (111) interface and its effect on photochemical reactivity

Yisi Zhu, Paul A. Salvador, and Gregory S. Rohrer* Carnegie Mellon University, Department of Materials Science and Engineering, Pittsburgh, PA 15213

Abstract High temperature annealing in air is used to produce SrTiO3 (111) surfaces with two types of atomically flat terraces: those that promote photoanodic reactions and those that promote photocathodic reactions. Surface potential measurements show that the photocathodic terraces have a relatively more positive surface potential than the photoanodic terraces. After depositing thin TiO2 films on the surface, from 1 nm to 13 nm thick, the surface of the film above the photocathodic terraces also has photocathodic properties, similar to those of the bare surface. While a more positive surface potential can be detected on the surface of the thinnest TiO2 films (1 nm thick), it is undetectable for thicker films. The persistence of the localized photocathodic properties on the film surface, even in the absence of a measurable difference in local potential, indicates that the charge associated with specific terraces on the bare SrTiO3 (111) surface remains localized at the TiO2/SrTiO3 interface and that the buried charge influences the motion of photogenerated carriers. *corresponding author email address: [email protected] KEYWORDS: SrTiO3, TiO2, Heterostructure, Photochemical Reactivity, Atomic Force Microscopy

ACS Paragon Plus Environment

ACS Applied Materials & Interfaces

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

1. Introduction Studies of polar surfaces such as ZnO (0001)1 and GaN (0001)2 have shown that the internal electric field that arises from the charge on the polar surface can separate photogenerated electron-hole pairs and increase their photochemical reactivity. SrTiO3 is a well-known photocatalyst that can split water when illuminated by UV light3-5 and several of the low index surfaces are polar. For example, the ideal unreconstructed (111) surface is polar and charged; it can be terminated by a Ti4+ layer or by a (SrO3)4- layer. While the real SrTiO3 (111) surface in air is likely to be reconstructed6-8 and covered by a layer of adsorbates,9, 10 it has been shown that some (111) terraces have locally photocathodic properties and the others have photoanodic properties.11 Furthermore, the photocathodic terraces have a relatively positive potential compared the photoanodic terraces, as measured by Kelvin probe force microscopy (KFM).12 The interpretation of these observations was that the terraces with a more positive potential (photocathodic) have reduced upward band bending at the surface and it is easier for photogenerated electrons to reach the surface and participate in photo-reduction reactions. The photoanodic terraces, with a less positive surface potential, have increased upward band bending that prevents electrons from getting to the surface while holes are driven to the surface to participate in oxidation reactions. In other words, the polar terraces on the SrTiO3 (111) surface have distinct photocathodic or photoanodic properties and, therefore, strongly influence photochemical reactions. While polar surfaces might provide a mechanism to increase photochemical reactivity, their stability with respect to faceting and other types of reconstruction is questionable.13, 14

One strategy for stabilizing polar surfaces is to add a thin coating that protects the polar

p. 2 ACS Paragon Plus Environment

Page 2 of 27

Page 3 of 27

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

surface.15-17 This may or may not be necessary for SrTiO3, whose native surfaces are relatively stable in photochemical reactions, especially compared to other complex oxides. However, the stabilities of SrTiO3 surfaces that have been processed to be more photocathodic or photoanodic, as in ref. 12, are not yet known. Earlier studies of ferroelectrics coated by TiO2 have illustrated that the reactivity of the film surface is spatially selective and mimics the reactivity of the underlying polar substrate.15, 17 The proposed mechanism is essentially the same as for the bare ferroelectric surface;18 the substrate surfaces with positively charged domains have reduced upward band bending and the substrate surfaces with negatively charged domains have increased upward band bending. Electrons and holes photogenerated in the substrate migrate to positive and negative domains, respectively, and can then pass through the relatively thin TiO2 coating and react with species at the surface. Recent simulations of photogenerated charge carrier transport in coated ferroelectrics provided quantitative support for this mechanism and suggested that further improvements can be made by the addition of electrocatalytic coatings.19, 20 Considering the similarities between charged ferroelectric domains and polar surface terminations, we can hypothesize that a thin TiO2 coating on a polar SrTiO3 surface will have, if the buried charge can be preserved at the interface, photocathodic and photoanodic regions that mimic the reactivity of the underlying substrate. The main point of this paper is to describe findings that support this hypothesis. TiO2 is a reasonable choice for the coating because it has been used to coat ferroelectrics and SrTiO3 in the past,15, 21 it is a stable photocatalyst,22, 23 and the TiO2/SrTiO3 heterostructure has been shown to be photocatalytically active.24 SrTiO3 (111) surfaces have been prepared by high temperature annealing and the photocathodic regions have been identified using both surface potential

p. 3 ACS Paragon Plus Environment

ACS Applied Materials & Interfaces

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

imaging by KFM and the photochemical reduction of silver. After growing thin TiO2 films, the same areas were analyzed and the results show that the charge on the terraces of the bare SrTiO3 surface remains at the buried interface and influence the transport of photogenerated charge carriers.

2. Experimental Procedure Commercially available SrTiO3 (111) single crystals (MTI company, Richmond, CA, (111) +/- 0.5°, roughness < 5 Å) were heated in air at 5 °C/min to 1250 °C and held for 10 h. After annealing, the topography and surface potential distribution were imaged using a scanning probe microscope (Solver-Next, NT-MDT, Russia). A silicon probe (Tap300-G, BudgetSensors, Bulgaria) was used to record topographic images and a conductive PtCr coated probe (190E-G, BudgetSensors, Bulgaria) was used to record surface potential (also referred to as KFM) images. The image data was processed using the Gwyddion25 software package. The photocathodic areas of the surface were identified by photochemically reducing Ag+ from solution.11, 18, 26, 27 The reduced silver is insoluble and forms metallic deposits on the surface that can be imaged by AFM.28 In this experiment, a Viton O-ring (diameter ~ 0.5 cm) is placed on the surface and a 0.115 M aqueous solution of AgNO3 was poured into the O-ring and a quartz slip was placed on top, sealing the solution in the O-ring by capillarity without an air bubble. The assembly was then illuminated with 300 W mercury lamp (Newport, Irvine, CA) for 8 s. During this procedure, skin and eyes must be protected to prevent damage from UV radiation. After illumination, the sample was rinsed with deionized water and dried using a stream of 99.995 % nitrogen gas. After imaging the pattern of silver on the surface, the silver was removed by wiping the surface with a cotton swab p. 4 ACS Paragon Plus Environment

Page 4 of 27

Page 5 of 27

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

and then sonicating the sample in a bath of methanol for 10 min. After the surface was cleaned, it was imaged by AFM again to make sure there was no detectable silver residue. TiO2 thin films were grown on the clean SrTiO3 (111) surface using pulsed laser deposition (PLD) (Neocera, MD) with a KrF (λ = 248 nm) laser (Coherent, CA). The films were deposited at 700 °C with an oxygen pressure of 30 to 40 mTorr. These conditions were adopted, with minor adjustments, from previous work in which TiO2 films were grown on BiFeO3 and BaTiO3 substrates.15, 17, 29 Note that to make the film as smooth as possible, we used a reduced laser pulse frequency (3 Hz) and a reduced laser energy (estimated to be 0.6 J/cm2 at the target). After deposition, the film thickness was measured using X-ray reflectivity. Films thicker than 5 nm can be measured directly, and the thicknesses of thinner films were estimated from the number of pulses used during the deposition (calibrated from the measurements on the thicker films). The film growth rate in these conditions is around 1 nm per 1400 pulses. The phase of the film was examined by X-ray diffraction. A Philips X’Pert Pro MRD system was used for the diffraction measurements. The Cu anode X-ray source was operated at 40 kV and 45 mA to generate an X-ray beam with a wavelength of 0.154 nm (Cu-Kα1/Kα2). Electron backscatter diffraction patterns of the heterostructure were captured using a Quanta 200 SEM (FEI, Hillsboro, OR) and analyzed using commercial software (TSL, EDAX, Mahwah, NJ) to further characterize the crystallinity and orientation of the films. The film surfaces were used to reduce silver under the same conditions as the bare substrates. Topographic atomic force microscopy (AFM) images were recorded of the same areas before and after depositing the film so that the patterns of silver reduced on the film could be compared to those on the bare substrate. KFM was also used to map the

p. 5 ACS Paragon Plus Environment

ACS Applied Materials & Interfaces

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

surface potential distribution in the same areas before and after film growth. It was possible to locate and image the same areas at different stages of the experiment by navigating the microscope with reference to visible fiducial marks deliberately added to the sample surface.

3. Results X-ray diffraction was used to characterize the bare annealed SrTiO3 (111) substrate and the 50 nm TiO2/SrTiO3 heterostructure. The diffraction patterns from the bare substrate and the heterostructures are shown in Fig. 1(a) and 1(b), respectively. When they are compared, it is clear that a new peak has appeared after deposition of the 50 nm film. The position the SrTiO3 (111) peak is 40.081°, shifted by +0.082° from the ideal peak position (39.999°).30 The film peak is located at 37.812°. Assuming that the measured position of the new peak is shifted by the same offset as the SrTiO3 (111) peak, its true position is 37.730°. We assign this new peak to anatase (004) because it differs from the ideal position by only 0.028°.31 It has been reported that TiO2 films grown on SrTiO3 (111) can have mixed phases and orientations: in addition to (001) anatase, (112) anatase and (100) rutile are also observed.32 However, the (112) anatase diffraction peak (38.600°) and the rutile (200) peak (39.195°) coincide with the SrTiO3 (111) substrate peak, so the X-ray diffraction data was not sufficient to determine if these other orientations exist.

p. 6 ACS Paragon Plus Environment

Page 6 of 27

Page 7 of 27

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

Figure 1. XRD patterns of (a) the bare SrTiO3 (111) substrate and (b) the TiO2/SrTiO3 heterostructure. The film thickness is about 50 nm. (c) An example of an X-ray reflectivity curve measured on a TiO2 film/SrTiO3 (111) single crystal heterostructure. (d) Electron backscatter diffraction pattern from the 50 nm thick TiO2 film on SrTiO3 (111).

A typical X-ray reflectivity pattern is shown in Fig. 1(c). The film thickness is directly proportional to the spacing between local maxima, while the overall intensity is also related to the substrate and film roughness. Using the X'Pert Reflectivity simulation software33, the pattern was simulated using a film (substrate) roughness of ≈ 4 Å (0 Å) and a film thickness of 7.5 nm. Note that the maximum incident angle range in the measurement is 4°; beyond 4° the intensity is less than the noise level. Therefore, the thinnest films that the X-ray reflectivity technique can characterize are around 5 nm, which will only show two maxima in the reflection intensity. Using the growth rate data from the thickest films, a series of films with thicknesses from 1 to 25.5 nm were deposited. An electron backscatter diffraction pattern of a 50 nm thick TiO2 film is shown in Fig. 1(d). The diffraction pattern shows some obvious bands, but is relatively weak, and could not be indexed to a single phase, though some bands (labeled in Fig. 1(d)) were consistent

p. 7 ACS Paragon Plus Environment

ACS Applied Materials & Interfaces

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

with the anatase (001) orientation. The overall pattern is best described as arising from several orientations or phases, indicating the film is polycrystalline. The above results indicate that the films are flat (supported later by AFM measurements), likely from kinetic factors including low relative adatom mobility. Furthermore, they have crystallites with small lateral spatial extents and several orientations, likely from specifics of the deposition conditions and the substrate surface itself, the former of which were targeted to yield smooth films and the latter of which were chosen to yield surfaces with specific photocatalytic properties. Such films are reasonable for our studies because polycrystalline films are characteristic of what would be expected in a real catalyst. Furthermore, previous studies of TiO2 coated BaTiO3 showed that the phase and orientation of the TiO2 had little effect on the reactivity, which was dominated by the substrate polarity.34 We expect the same to be true here. The topography of the SrTiO3 (111) surface after annealing at 1250 °C in air is shown in Fig. 2(a). The surface is characterized by a set of atomically flat terraces separated by approximately parallel steps 1 to 2 nm in height. The roughness of a single terrace on the SrTiO3 (111) substrate is < 0.02 nm, much less than the height between parallel layers, and we therefore refer to the terraces as “atomically flat”. The surface potential distribution in the same area is shown in Fig. 2(b). The surface potential image indicates that the terraces have two distinct contrast levels and the potential difference between the two varies in the range of 7 to 15 mV. After the sample was illuminated by UV light in a silver nitrate solution, metallic silver selectively deposited on some of the terraces and appears as white contrast in Fig. 2(c); the deposits have a height of approximately 5 nm. After comparing the locations of the silver to the potential distribution in the KFM image (Fig. 2(b)), it is clear the Ag deposits preferentially on the set of terraces that have the relatively greater (more

p. 8 ACS Paragon Plus Environment

Page 8 of 27

Page 9 of 27

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

positive) surface potential. These observations are consistent with those we have recently reported and we refer to the terraces that reduce silver as photocathodic terraces.12

Figure 2. (a,d) Topographic AFM images and (b,e) surface potential images before the reaction, and (c,f) topographic images of the same areas after the photochemical reduction of Ag+. (a-c) Images of a bare SrTiO3 surface and (d-f) a 1 nm thick TiO2 film surface supported on the same SrTiO3 (111) substrate at the same location. The field of view for all images is 5 µm × 5 µm. The coloring scheme used in all images in this paper is illustrated between panels (a) and (b), where black (white) is assigned to the lowest (highest) numerical value. In these images, the black-to-white contrast range is: (a, d) 0 – 5 nm, (b) 0 – 30 mV, (e) 0 – 20 mV, and (c, f) 0 – 10 nm.

The reduced silver was removed from the surface and a 1 nm thick TiO2 film was deposited. A topographic AFM image (Fig. 2(d)) and a KFM image (Fig. 2(e)) were recorded of the film surface at the same location where the images in Fig. 2(a-c) were recorded. There is no obvious topographic change after film deposition and the roughness on the terrace is approximately 0.04 nm, indicating that the film is also atomically flat on p. 9 ACS Paragon Plus Environment

ACS Applied Materials & Interfaces

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

each terrace. The potential contrast measured on the film surface (Fig. 2(e)) is similar to that measured on substrate (Fig. 2(b)), though the measured potential differences between types of terraces for the film (3 to 8 mV) are less than those for the substrate (7 to 15 mV). The topography of the same area of the film’s surface after the photoreduction of silver is shown in Fig. 2(f). The pattern of silver on the film surface is similar to that observed on the substrate surface before film deposition. In the case of the film, some silver deposits appear everywhere on the film, but significantly more silver is observed on the terraces with a higher potential. In other words, regions above the photocathodic terraces on the bare substrate are also photocathodic on the surface of the covering film. Additional images of this sample with a 15 μm × 15 μm field of view, illustrating the same features in Fig. 2, are displayed in Fig. S1 (supporting information). The experiment described above was repeated, but using a thicker, 7.5 nm TiO2 film, and the results are illustrated in Fig. 3. The images in Figs. 3(a), (b), and (c) illustrate the topography of the clean substrate, the surface potential, and the topography after the photochemical reduction of silver, respectively. The images in Fig. 3(a-c) have the same characteristics as those in Fig. 2(a-c). Specifically, the measured surface potential of the atomically flat terraces is correlated to the location of the photochemcially reduced silver. After removing the silver and depositing a 7.5 nm thick TiO2 film, images of the same area were recorded before (Fig. 3(d) and (e)) and after (Fig. 3(f)) the photochemical reduction of silver. Comparing Fig. 3(a) and (d), the film is obviously rougher than the substrate, but the shapes of the step edges and terraces are unchanged. Thus, it is possible to verify that the image is recorded at the same location. The KFM image contrast (Fig. 3(e)) of the film is not similar to that of the substrate (Fig. 3(e)) and it thus differs from the corresponding

p. 10 ACS Paragon Plus Environment

Page 10 of 27

Page 11 of 27

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

image of the 1 nm film (Fig. 2(e)). The potential contrast in Fig. 3(e) may be related to the polycrystalline nature of film, as different TiO2 phases and orientations have different work functions, which contribute to the local surface potential. Even though the potential distribution on the substrate is not visible in the KFM image, the pattern of reduced silver (Fig. 3(f)) still mimics that found on the bare substrate (Fig. 3(c)). This result suggests that the charge on the buried substrate terraces remains and influences the motion of photogenerated carriers, but is screened by the film so that it is not detected by the KFM tip which is above the surface.

Figure 3. (a,d) Topographic AFM images and (b,e) surface potential images before the reaction, and (c,f) topographic images of the same areas after the photochemical reduction of Ag+. (a-c) Images of the bare SrTiO3 surface and (d-f) a 7.5 nm thick TiO2 film surface supported on the same SrTiO3 (111) substrate at the same location. The field of view for all images is 5 µm × 5 µm. In these images, the black-to-white contrast range is: (a, d) 0 – 5 nm, (b) 0 – 25 mV, (e) 0 – 20 mV, and (c, f) 0 – 10 nm.

p. 11 ACS Paragon Plus Environment

ACS Applied Materials & Interfaces

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Additional images of this sample with a 15 μm × 15 μm field of view, illustrating the same features in Fig. 3, are displayed in Fig. S3 (supporting information). The experiments described above, using 1 nm and 7.5 nm thick TiO2 films, were also carried out using samples with 2 nm and 13 nm thick TiO2 films; the results from these experiments are shown in Fig. S2 and Fig. S4, respectively, in the supporting information. The results from the 2 nm and 13 nm thick TiO2 films are similar to what was observed for the 7.5 nm thick film. Specifically, the potential contrast in the KFM images of the substrate does not appear in KFM images of the film, but the pattern of reduced silver on the film surface is similar to what was observed on the substrate surface before film deposition. In the crystal structure of SrTiO3, neighboring Ti4+ and (SrO3)4- (111) layers are separated by N = 1.12 Å. Terraces separated by an even-N step height are of the same surface composition while terraces separated by an odd-N step height have different surface composition. These step height differences have also been demonstrated experimentally, and have been shown to correlate with photochemical reactivity and surface potential measurements.11, 12 The height and surface potential profile measured on the bare substrate along the (blue) line drawn on Fig. 2(a) and (b) are plotted in Fig. 4(a), respectively using a black dashed line and light grey solid line. The average heights of the photocathodic terraces are marked with solid red lines. The other terraces, shown to be photoanodic in previous work,11, 12 are marked with solid blue lines. From left to right, the line traverses six terraces. We define the height of 1st terrace as 0 Å, for reference, and determine the heights (number of N layers) of the 2nd to 6th terraces as 5.7 Å (5N), 25.7 Å (23N), 33.4 Å (30N), 42.2 Å (38N), and 65.7 Å (59N). Note that the N in parentheses corresponds to the difference from the reference at 0 Å, while the N marked in the Fig. 4

p. 12 ACS Paragon Plus Environment

Page 12 of 27

Page 13 of 27

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

corresponds to the height difference between neighboring terraces. After correlating the step heights with the surface potential profile, we find all photocathodic terraces have higher surface potentials (18 to 22 mV) and, because they are separated by even-N step heights, have the same surface composition. The remaining non-photocathodic terraces have lower surface potentials (11 to 13.5 mV) and, because they are separated by even-N step heights, have the same surface composition.

Figure 4. Surface potential and height measurements extracted from images in Figs. 2 and 3 at the positions of the blue lines; (a) the substrate and the 1 nm film and (b) the substrate and the 7.5 nm film. The black dashed lines and light grey solid lines are the height and surface potential profiles from the bare SrTiO3 substrates, respectively. The black solid lines are the height profiles from the same locations on the film surfaces. The average heights of photocathodic (photoanodic) terraces are marked with red (blue) lines in the image. All step heights are integer multiples of the inter-planar spacing N (N =1.12 Å).

p. 13 ACS Paragon Plus Environment

ACS Applied Materials & Interfaces

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The height profile extracted from the film surface at exactly the same location is plotted with a solid black line in Fig. 4(a). There is no significant change for the step heights measured on the film surface, indicating the films are flat and the growth rate did not depend on the composition of the terrace. The surface roughnesses of single terraces, represented by the standard deviation of the height, were also calculated from these data. For the bare SrTiO3 substrate, the roughnesses of single terraces were between 0.1 and 0.2 Å, while the roughnesses of the film terraces were between 0.4 and 0.5 Å; therefore, the 1 nm thick film is considered atomically flat and the substrate terrace structures appear to have been well preserved at the film surface. The results from the substrate and 7.5 nm thick film, extracted from the positions of the blue lines in the images in Fig. 3, are illustrated in Fig. 4(b). Again for the substrate surface, the photocathodic terraces have high surface potentials (17.5 to 20 mV) and other terraces have low potentials (11.5 to 14 mV). Also, the changes in the potential can be correlated with step heights that change the termination. The line traverses six terraces and, from left to right, their heights are 0 Å, 14.7 Å (13N), 20.2 Å (18N), 32.6 Å (29N), 38.1 Å (34N) and 51.7 Å (46N). The photocathodic terraces are separated by even-N steps which means they have the same composition, and they are separated from the non-photocathodic terraces by odd-N steps, which therefore have a different composition. The surface roughness of the substrate terraces was also between 0.1 Å and 0.2 Å. The roughnesses of terraces on the film were ≈ 1.4 Å, but this does not affect the patterns of reduced silver. The experiments carried out on the 1 nm (Fig. 2), the 2 nm (Fig. S2), 7.5 nm (Fig. 3), and the 13 nm thick TiO2 film (Fig. S4) were repeated using a heterostructure consisting of a SrTiO3 (111) substrate and a 25.5 nm thick film. The patterns of silver reduced on the

p. 14 ACS Paragon Plus Environment

Page 14 of 27

Page 15 of 27

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

surface of the SrTiO3 (111) substrate and on the surface of the 25.5 nm film at the same location are shown in Fig. 5(a) and (b), respectively (note the images in Fig. 5 is 15 x 15 µm2). While the 25.5 nm thick film conseals many of the substrate features, it is still possible to locate the traces of large steps and surface scratches. The arrows at the bottom of the images locate equivalent positions on the substrate and film surface. As for all of the substrates, photochemically reduced silver was only found on the terraces that had a higher surface potential (the corresponding KFM image is presented in Fig. S5 of the supporting information). The pattern of reduced silver on the 25.5 nm thick film does not correlate well with the pattern observed on substrate; instead, there are silver particles relatively evenly distributed across the surface. This result implies that the photochemical reactivity of thicker films (> than 25 nm) is not dominated by the potential difference at the buried substrate-film interface, in contrast to the results for films of thicknesses 1 to 13 nm.

Figure 5. Topographic AFM images of (a) a bare SrTiO3 (111) substrate and the same areas of (b) a 25.5 nm TiO2 film on the substrate after the photoreduction of Ag+. Both images are 15 μm × 15 μm and the black-to-white contrast range is 0 – 15 nm. The arrows mark equivalent locations.

p. 15 ACS Paragon Plus Environment

ACS Applied Materials & Interfaces

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

4. Discussion The observations in the previous section show that TiO2 films (up to 13 nm in thickness) supported on SrTiO3 (111) surfaces (which were annealed in air at 1250 °C for 10 h prior to growth) have roughly the same spatially selective reactivity as the bare substrates. For the thinnest film, the relative potential difference between the two terraces is also similar to the potential difference on the bare substrate. These observations indicate that the substrate surface charge is stable with respect to TiO2 film growth. The charge on the terraces of the bare substrate is trapped at the buried substrate-film interface where it continues to influence the transport of photogenerated electrons and holes. To understand the mechanism of this phenomenon, we begin by considering where the photoelectrons are generated. SrTiO3 and TiO2 are both able to reduce Ag when stimulated by UV irradiation,11, 26 but in this experiment the TiO2 film is too thin to absorb enough light to produce significant concentrations of electron-hole pairs. Because the light used in this experiment has an absorption depth in SrTiO3 and TiO2 on the order of 102 nm,35-37 the majority of light was adsorbed in the substrate; most of the electrons participating in the silver-reduction reaction must also have been generated in the SrTiO3 substrate. Next we consider the diffusion length of electrons in TiO2, which is on the order of 102 nm.38 Because the TiO2 films were all less than 13 nm thick, the photoelectrons from the substrate can easily get to the film surface by diffusion. Note that because the average terrace width is more than ten times the film thickness, we assume that carrier diffusion perpendicular to the film influences the reaction more than lateral diffusion. Next we

p. 16 ACS Paragon Plus Environment

Page 16 of 27

Page 17 of 27

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

consider the buried charge. The KFM images measured on the ultra-thin film sample show that, after we deposit a 1 nm TiO2 film on the annealed SrTiO3 (111) substrate, there is still a correlation between surface potential and reactivity, similar to the surface potential and reactivity correlation for bare SrTiO3 (111) substrates. This suggests that the potential difference on the film surface originates from the potential differences of the photocathodic and photoanodic terraces on the buried SrTiO3 (111) surface. Taken together, these considerations indicate that carriers photo-generated in the substrate experience a field from the buried interface charge that attracts different amounts of electrons and holes to the interface, depending on the buried local terrace charge, and some of these carriers traverse the film and react at the TiO2 surface. Schematic energy level diagrams are depicted in Fig. 6 for (a) the bare SrTiO3 (111) substrate, (b) an ultra-thin TiO2 film (≈ 1 nm) on SrTiO3 (111), and (c) thin (1 – 13 nm) TiO2 films on SrTiO3 (111), all grounded with a PtCr coated AFM probe. (Schematic energy level diagrams for SrTiO3, anatase TiO2, and the PtCr coated AFM probe before contact can be found in Fig. S6 in the supporting information.) Both SrTiO3 and TiO2 are n-type semiconductors, likely owing to oxygen vacancies.39-41 In the case of a negative (positive) surface or interfacial charge, shown by blue (red) bands, the bands are bent upward (downward) and modify the barrier (field) for electron drift. In other words, the oppositely charged terraces bend the bands in opposite directions. The more (less) negatively charged terraces will have a larger (smaller) barrier to electron drift to the surface / interface, and are expected to be less (more) photocathodic. For the TiO2/SrTiO3 heterostructures, the electrons must traverse the thin TiO2 layer, either by drift or diffusion, to participate in surface reactions. This has been shown in recent simulations to

p. 17 ACS Paragon Plus Environment

ACS Applied Materials & Interfaces

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

have a relatively small influence on the overall reactivity of a terrace.19, 20 In other words, the buried chemical charge of the SrTiO3 surface, which can be controlled through annealing,12 behaves schematically similar to the buried polar surfaces in ferroelectric heterostructures.

Figure 6. Schematic energy level diagrams of (a) SrTiO3 (111), (b) ultra-thin (1 nm) TiO2 on SrTiO3 (111) and, (c) thin TiO2 (1 – 13 nm) on SrTiO3 (111), each grounded with PtCr coated conductive AFM tips. In each, the Fermi levels (EF) are aligned in all phases while the reference level (ERef) discontinuities between the sample surface and tip represent the contact potential differences (CPDs) as would be measured by KFM. In each panel, the blue (red) arrows indicate the CPD for photoanodic (photocathodic) terraces. The band bending in the SrTiO3 is affected by the charge on its surface, regardless of whether a film is located on it or not, and controls reactivity in each case.

The surface potential value measured by KFM is proportional to the contact potential difference (CPD) between the probe and sample surface,42, 43 and is shown in the schematics of Fig. 6 as the discontinuity in the vacuum levels. Eref, E-, and E+ denote the vacuum level positions for the AFM tip, photoanodic terraces, and photocathodic terraces, respectively. A negative (positive) surface with upward (downward) band bending results in a smaller (larger) value of the CPD, or the energy difference between Eref and E- (E+), and a smaller (larger) measured surface potential. The potential difference between

p. 18 ACS Paragon Plus Environment

Page 18 of 27

Page 19 of 27

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

photocathodic and photoanodic terraces is represented as ∆E in Fig. 6. For the charged bare substrate surfaces, Fig. 6(a), and the ultra-thin film on the charged substrate surfaces, Fig. 6(b), where the film is not thick enough to fully screen the substrate charge, one can measure a surface potential difference between the terraces, though the values are smaller for the film than for the bare substrate. As the film thickness is increased, the buried charge is better screened by the film and the band edges for the films are closer to their equilibrium positions, as shown in Fig. 6(c). In such cases, the CPD (and surface potential) measured for the film above either type of substrate terrace will be similar. It should be noted that, in Fig. 6(c), the film is thinner than the space charge region needed to screen the interface charge originating from SrTiO3 polar terminations (thus the TiO2 bands at the TiO2/PtCr interface are not bent upward as they would be for bulk n-type TiO2).19, 20 The experiments can be consistently interpreted according to the schematics in Fig. 6, with the substrates similar to (a), the 1 nm film similar to (b), and films over 2 nm similar to (c), even though the exact amount of band bending and the relative differences are not known. The observation that the spatially selective photochemical reactivity of the substrate can be transferred to a film that normally exhibits spatially uniform reactivity is not completely new. The same phenomenon was observed in TiO2/BaTiO3 and TiO2/BiFeO3 heterostructures.15, 17, 34 However, while both BaTiO3 and BiFeO3 are ferroelectric, SrTiO3 is not. The observation of this phenomenon in a non-ferroelectric material means that it might occur for almost any compound with significant iconicity, rather than only for a limited number of ferroelectrics. For the cases of BaTiO3 and BiFeO3, ferroelectric domains in the bulk with a positive out-of-plane polarization at the surface are photocathodic, while domains with a negative surface polarization are photoanodic.18 Therefore, the patterns of

p. 19 ACS Paragon Plus Environment

ACS Applied Materials & Interfaces

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

reduced silver on BaTiO3 and BiFeO3 are correlated with the domain structure. Burbure et al.15 and Zhang et al.17 found that thin TiO2 films supported on BaTiO3 or BiFeO2 substrates photo-reduce silver in spatially selective patterns; the patterns of the silver were correlated with the substrate domain structure. The phenomenon is similar for the TiO2/SrTiO3 heterostructure, except that SrTiO3 is not ferroelectric. We therefore assert that it is the potentials of the polar surface terminations on different buried terraces that influence the motion of photogenerated carriers. In this case, the phenomenon is quite different from a ferroelectric. In the ferroelectric, the polarization arises from the bulk domain structure. Here, it is from the composition of the last layer of the crystal. While it is not surprising that the addition of a film does not alter the bulk domain structure of a ferroelectric, it is more surprising that the growth a film does not significantly alter the charge at the surface of the substrate. The reason for this is that the TiO2 film is not grown on an ideal SrTiO3 (111) surface. The high temperature annealing that was used to prepare charged terraces on the substrate is likely to create a reconstructed surface region (that may extend several layers below the surface) with varying degrees of Sr deficiency departing from the ideal composition.12 The film growth occurs at a much lower temperature (700 °C) and does not significantly alter the structure and composition of the SrTiO3 surface layer formed at the higher temperature (1250 °C). So, rather than growing on an ideal SrTiO3 substrate, the films examined here grew on a reconstructed, Sr-defficient surface layer. The fact that the TiO2 films that grow on this surface are polycrystalline and multiphased supports the idea that the high temperature anneal creates a non-ideal growth surface on the substrate. While non-ideal for epitaxy, the reconstructed surfaces are stable against growth and their charges are important for the photochemical properties of the heterostructure.

p. 20 ACS Paragon Plus Environment

Page 20 of 27

Page 21 of 27

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

For TiO2 films supported on BaTiO3 substrates, the buried ferroelectric domains no longer controlled the photochemical reactivity of film surfaces when the film thickness was beyond 50 nm.15 A computational model of the TiO2/BaTiO3 heterostructure found that the electron and hole current densities remain the same when the film thickness is beyond ≈ 30 nm, indicating that the photocarriers arriving at film surface are mainly generated in the film beyond this thickness.19, 20 These observations agree with the observation in this paper that when the thickness of TiO2 film is increased (here to 25.5 nm), the photochemical reactivity of films supported on photocathodic and photoanodic terraces become similar, as more photoelectrons are photogenerated in the TiO2 layer. The diffusion length of electrons in TiO2 is greater than 100 nm,38 though the diffusion length in polycrystalline films should smaller than this. Some combination between decreased driving forces and small diffusion lengths leads to the disappearance of substrate induced spatial selectivity at thicknesses between 13 nm and 25.5 nm. It should be noted that earlier studies showed that TiO2/SrTiO3 composites had enhanced UV light photocatalytic activity compared to the separated phases.21, 24 In the TiO2/SrTiO3 heterostructures described here, we also observed slightly enhanced photoreactivity for the reduction of silver. Films supported on originally unreactive SrTiO3 terraces (photoanodic) reduced some silver, while films supported on photocathodic SrTiO3 terraces were just as reactive as the bare surface. This is understandable because TiO2 is a good photocatalyst when activated by UV light. Although the film is thin, it still absorbs some light and produces photocarriers to participate in the reaction. It should also be noted that the electronic structure of the heterostructure can promote the separation of photogenerated charge carriers. Because both the conduction band and valence band of SrTiO3 are at higher energies than the comparable bands in TiO2 (as depicted in Fig. 6 and

p. 21 ACS Paragon Plus Environment

ACS Applied Materials & Interfaces

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Fig. S6)44-46 photogenerated electrons in SrTiO3 can reduce their energy by moving to the conduction band of TiO2 and ultimately to the surface to reduce Ag+ from solution. The findings reported here might ultimately have an important impact on the design of photocatalysts. The idea of using fixed charges on catalyst surfaces to separate photogenerated carriers, ameliorate recombination, and increase reaction efficiency is well established.47-51 By burying these charges at an internal interface covered by TiO2, the catalyst will be stable in an aqueous environment and the charges will be protected from degradation. While this has been demonstrated before for TiO2 coated ferroelectrics,15, 17, 21 SrTiO3 offers a significant possible advantage compared to ferroelectrics because the ratio of photocathodic to photoanodic surface area can be controlled.12 The relative areas of the polar domains on the surface of a ferroelectric are nearly equal. The best ratio for the areas of the photocathodic and photoanodic surfaces is the one that maximizes the overall reaction rate. This is presumably the inverse of the ratio of the area specific rates of the cathodic and anodic half reactions, and this is unlikely to be unity. While the ratios of opposite domains can be controlled in bulk ferroelectrics by poling, it is not clear how this could be accomplished with a powdered catalyst. However, we have recently demonstrated that the ratio of photocathodic and photoanodic surface area on the SrTiO3 (111) surface can be controlled by high temperature thermal treatments.12 Therefore, it is possible, at least in principle, to create catalysts with controlled distributions of surfaces promoting photocathodic or photoanodic reactions and to protect these surfaces with a thin TiO2 layer that also preserves the benefits associated with charge carrier separation.

p. 22 ACS Paragon Plus Environment

Page 22 of 27

Page 23 of 27

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

5. Conclusions When silver is photochemically reduced on thin (≤ 13 nm) TiO2 films supported by SrTiO3 (111) substrates, the pattern of silver on the film surface mimics the pattern that is reduced on the substrate surface, before the film is deposited. Silver is preferentially reduced above terraces that have a more positive surface potential than the unreactive terraces. These observations indicate that the charges on different terraces on the SrTiO3 (111) surface remain at the buried interface after film growth. At the interface, they influence the transport of photogenerated charge carriers to the surface such that silver is reduced on the film above the photocathodic terraces on the substrate.

Supporting Information. The supporting information includes additional scanning probe microscopy data from 1, 2, 7.5, 13, and 25.5, nm thick films and proposed energy level diagrams for the heterostructures. Acknowledgements The authors acknowledge the support of National Science Foundation grant DMR 1609369 and use of the Materials Characterization Facility at Carnegie Mellon University supported by grant MCF-677785.

References Cited 1. Yang, J.; Wang, J.; Li, X.; Lang, J.; Liu, F.; Yang, L.; Zhai, H.; Gao, M.; Zhao, X., Effect of Polar and Non-Polar Surfaces of ZnO Nanostructures on Photocatalytic Properties. J. Alloys Compd. 2012, 528, 28-33. 2. Bae, H.; Kim, E.; Park, J.-B.; Kang, S.-J.; Fujii, K.; Lee, S. H.; Lee, H.-J.; Ha, J.-S., Effect of Polarity on Photoelectrochemical Properties of Polar and Semipolar GaN Photoanode. J. Electrochem. Soc. 2016, 163, H213-H217. 3. Lehn, J. M.; Sauvage, J. P.; Ziessel, R., Photochemical Water Splitting Continuous Generation of Hydrogen and Oxygen by Irradiation of Aqueous Suspensions of Metal Loaded Strontium-Titanate. New J. Chem. 1980, 4, 623-627. p. 23 ACS Paragon Plus Environment

ACS Applied Materials & Interfaces

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

4. Luo, J. H.; Maggard, P. A., Hydrothermal Synthesis and Photocatalytic Activities of SrTiO3-Coated Fe2O3 and BiFeO3. Adv. Mater. 2006, 18, 514-517. 5. Townsend, T. K.; Browning, N. D.; Osterloh, F. E., Nanoscale Strontium Titanate Photocatalysts for Overall Water Splitting. ACS Nano 2012, 6, 7420-7426. 6. Chiaramonti, A. N.; Lanier, C. H.; Marks, L. D.; Stair, P. C., Time, Temperature, and Oxygen Partial Pressure-Dependent Surface Reconstructions on SrTiO3(111): A Systematic Study of Oxygen-Rich Conditions. Surf. Sci. 2008, 602, 3018-3025. 7. Marks, L. D.; Chiaramonti, A. N.; Tran, F.; Blaha, P., The Small Unit Cell Reconstructions of SrTiO3(111). Surf. Sci. 2009, 603, 2179-2187. 8. Russell, B. C.; Castell, M. R., Surface of Sputtered and Annealed Polar SrTiO3(111): TiOx-Rich (N X N) Reconstructions. J. Phys. Chem. C 2008, 112, 6538-6545. 9. Ferrer, S.; Somorjai, G. A., UPS and XPS Studies of Chemisorption of O2, H2, and H2O on Reduced and Stoichiometric SrTiO3(111) Surfaces - the Effects of Illumination. Surf. Sci. 1980, 94, 41-56. 10. Nagarkar, P. V.; Searson, P. C.; Gealy, F. D., Effect of Surface-Treatment on SrTiO3 - an X-Ray Photoelectron Spectroscopic Study. J. Appl. Phys. 1991, 69, 459-462. 11. Giocondi, J. L.; Rohrer, G. S., Structure Sensitivity of Photochemical Oxidation and Reduction Reactions on SrTiO3 Surfaces. J. Am. Ceram. Soc. 2003, 86, 1182-1189. 12. Zhu, Y.; Salvador, P. A.; Rohrer, G. S., Controlling the Relative Areas of Photocathodic and Photoanodic Terraces on the SrTiO3(111) Surface. Chem. Mater. 2016, 28, 5155-5162. 13. Goniakowski, J.; Finocchi, F.; Noguera, C., Polarity of Oxide Surfaces and Nanostructures. Rep. Prog. Phys. 2008, 71, 016501 (55pp). 14. Tasker, P. W., Stability of Ionic-Crystal Surfaces. J. Phys. C: Solid State Phys. 1979, 12, 4977-4984. 15. Burbure, N. V.; Salvador, P. A.; Rohrer, G. S., Photochemical Reactivity of Titania Films on BaTiO3 Substrates: Origin of Spatial Selectivity. Chem. Mater. 2010, 22, 58235830. 16. Heller, A., Chemistry and Applications of Photocatalytic Oxidation of Thin Organic Films. Acc. Chem. Res. 1995, 28, 503-508. 17. Zhang, Y. L.; Schultz, A. M.; Salvador, P. A.; Rohrer, G. S., Spatially Selective Visible Light Photocatalytic Activity of TiO2/BiFeO3 Heterostructures. J. Mater. Chem. 2011, 21, 4168-4174. 18. Giocondi, J. L.; Rohrer, G. S., Spatial Separation of Photochemical Oxidation and Reduction Reactions on the Surface of Ferroelectric BaTiO3. J. Phys. Chem. B 2001, 105, 8275-8277. 19. Glickstein, J. J.; Salvador, P. A.; Rohrer, G. S., Computational Model of Domain-Specific Reactivity on Coated Ferroelectric Photocatalysts. J. Phys. Chem. C 2016, 120, 12673-12684. 20. Glickstein, J. J.; Salvador, P. A.; Rohrer, G. S., Multidomain Simulations of Coated Ferroelectrics Exhibiting Spatially Selective Photocatalytic Activity with High Internal Quantum Efficiencies. J. Mater. Chem. A 2016, 4, 16085-16093. 21. Li, L.; Rohrer, G. S.; Salvador, P. A., Heterostructured Ceramic Powders for Photocatalytic Hydrogen Production: Nanostructured TiO2 Shells Surrounding Microcrystalline (Ba,Sr)TiO3 Cores. J. Am. Ceram. Soc. 2012, 95, 1414-1420. 22. Hirano, M.; Nakahara, C.; Ota, K.; Inagaki, M., Direct Formation of Zirconia-Doped Titania with Stable Anatase-Type Structure by Thermal Hydrolysis. J. Am. Ceram. Soc. 2002, 85, 1333-1335. p. 24 ACS Paragon Plus Environment

Page 24 of 27

Page 25 of 27

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

23. Inagaki, M.; Nakazawa, Y.; Hirano, M.; Kobayashi, Y.; Toyoda, M., Preparation of Stable Anatase-Type TiO2 and Its Photocatalytic Performance. Int. J. Inorg. Mater. 2001, 3, 809-811. 24. Zhang, J.; Bang, J. H.; Tang, C.; Kamat, P. V., Tailored TiO2-SrTiO3 Heterostructure Nanotube Arrays for Improved Photoelectrochemical Performance. ACS Nano 2010, 4, 387395. 25. Necas, D.; Klapetek, P., Gwyddion: An Open-Source Software for SPM Data Analysis. Cent. Eur. J. Phys. 2012, 10, 181-188. 26. Lowekamp, J. B.; Rohrer, G. S.; Hotsenpiller, P. A. M.; Bolt, J. D.; Farneth, W. E., Anisotropic Photochemical Reactivity of Bulk TiO2 Crystals. J. Phys. Chem. B 1998, 102, 7323-7327. 27. Munprom, R.; Salvador, P. A.; Rohrer, G. S., Polar Domains at the Surface of Centrosymmetric BiVO4. Chem. Mater. 2014, 26, 2774-2776. 28. Herrmann, J. M.; Disdier, J.; Pichat, P., Photocatalytic Deposition of Silver on Powder Titania - Consequences for the Recovery of Silver. J. Catal. 1988, 113, 72-81. 29. Burbure, N. V.; Salvador, P. A.; Rohrer, G. S., Orientation and Phase Relationships between Titania Films and Polycrystalline BaTiO3 Substrates as Determined by Electron Backscatter Diffraction Mapping. J. Am. Ceram. Soc. 2010, 93, 2530-2533. 30. Wyckoff, R. W. G., Crystal Structures. Interscience: New York, 1964; Vol. 2. 31. Cromer, D. T.; Herrington, K., The Structures of Anatase and Rutile. J. Am. Chem. Soc. 1955, 77, 4708-4709. 32. Lotnyk, A.; Senz, S.; Hesse, D., Epitaxial Growth of TiO2 Thin Films on SrTiO3, LaAlO3 and Yttria-Stabilized Zirconia Substrates by Electron Beam Evaporation. Thin Solid Films 2007, 515, 3439-3447. 33. PANalytical X'pert Pro Mrd Xl: Pw3217 X'pert Reflectivity, The Netherlands 2007. 34. Burbure, N. V.; Salvador, P. A.; Rohrer, G. S., Photochemical Reactivity of Titania Films on BaTiO3 Substrates: Influence of Titania Phase and Orientation. Chem. Mater. 2010, 22, 5831-5837. 35. Frye, A.; French, R. H.; Bonnell, D. A., Optical Properties and Electronic Structure of Oxidized and Reduced Single-Crystal Strontium Titanate. Z. Metallkd. 2003, 94, 226-232. 36. Sekiya, T.; Ichimura, K.; Igarashi, M.; Kurita, S., Absorption Spectra of Anatase TiO2 Single Crystals Heat-Treated under Oxygen Atmosphere. J. Phys. Chem. Solids 2000, 61, 1237-1242. 37. Tang, H.; Levy, F.; Berger, H.; Schmid, P. E., Urbach Tail of Anatase TiO2. Phys. Rev. B 1995, 52, 7771-7774. 38. Yamada, Y.; Kanemitsu, Y., Determination of Electron and Hole Lifetimes of Rutile and Anatase TiO2 Single Crystals. Appl. Phys. Lett. 2012, 101, 133907 (4pp). 39. Breckenridge, R. G.; Hosler, W. R., Electrical Properties of Titanium Dioxide Semiconductors. Phys. Rev. 1953, 91, 793-802. 40. Forro, L.; Chauvet, O.; Emin, D.; Zuppiroli, L.; Berger, H.; Levy, F., High-Mobility nType Charge-Carriers in Large Single-Crystals of Anatase (TiO2). J. Appl. Phys. 1994, 75, 633-635. 41. Robertson, J.; Chen, C. W., Schottky Barrier Heights of Tantalum Oxide, Barium Strontium Titanate, Lead Titanate, and Strontium Bismuth Tantalate. Appl. Phys. Lett. 1999, 74, 1168-1170.

p. 25 ACS Paragon Plus Environment

ACS Applied Materials & Interfaces

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

42. Melitz, W.; Shen, J.; Kummel, A. C.; Lee, S., Kelvin Probe Force Microscopy and Its Application. Surf. Sci. Rep. 2011, 66, 1-27. 43. Nonnenmacher, M.; Oboyle, M. P.; Wickramasinghe, H. K., Kelvin Probe Force Microscopy. Appl. Phys. Lett. 1991, 58, 2921-2923. 44. Pan, J.; Liu, G.; Lu, G. Q.; Cheng, H.-M., On the True Photoreactivity Order of {001}, {010}, and {101} Facets of Anatase TiO2 Crystals. Angew. Chem. Int. Ed. 2011, 50, 21332137. 45. Thomas, A. G.; Flavell, W. R.; Mallick, A. K.; Kumarasinghe, A. R.; Tsoutsou, D.; Khan, N.; Chatwin, C.; Rayner, S.; Smith, G. C.; Stockbauer, R. L.; Warren, S.; Johal, T. K.; Patel, S.; Holland, D.; Taleb, A.; Wiame, F., Comparison of the Electronic Structure of Anatase and Rutile TiO2 Single-Crystal Surfaces Using Resonant Photoemission and X-Ray Absorption Spectroscopy. Phys. Rev. B 2007, 75, 035105. 46. van Benthem, K.; Elsasser, C.; French, R. H., Bulk Electronic Structure of SrTiO3: Experiment and Theory. J. Appl. Phys. 2001, 90, 6156-6164. 47. Batzill, M., Fundamental Aspects of Surface Engineering of Transition Metal Oxide Photocatalysts. Energy Environ. Sci. 2011, 4, 3275-3286. 48. Kakekhani, A.; Ismail-Beigi, S., Ferroelectric-Based Catalysis: Switchable Surface Chemistry. ACS Catal. 2015, 5, 4537-4545. 49. Khan, M. A.; Nadeem, M. A.; Idrissn, H., Ferroelectric Polarization Effect on Surface Chemistry and Photo-Catalytic Activity: A Review. Surf. Sci. Rep. 2016, 71, 1-31. 50. Li, L.; Salvador, P. A.; Rohrer, G. S., Photocatalysts with Internal Electric Fields. Nanoscale 2014, 6, 24-42. 51. Tiwari, D.; Dunn, S., Photochemistry on a Polarisable Semi-Conductor: What Do We Understand Today? J. Mater. Sci. 2009, 44, 5063-5079.

p. 26 ACS Paragon Plus Environment

Page 26 of 27

Page 27 of 27

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

Table of Contents Graphic

p. 27 ACS Paragon Plus Environment